Mobile Phones for Agricultural and Rural Development - School of ...

wrendressMobile - Wireless

Nov 12, 2013 (3 years and 4 months ago)

55 views

Development Informatics


Working Paper Series


The Development Informatics working paper series discusses the broad issues
surrounding information, knowledge, information systems, and information and
communication technologies in the process of socio
-
econo
mic development






Paper No. 50


Mobile Phones for
Agricultural and Rural
Development:

A Literature
Review and Future Research
Directions


RICHARD DUNCOMBE



2012





ISBN:
978
-
1
-
905469
-
31
-
4


Published
by:

Centre for Development Informatics

Institute f
or Development Policy and Management, SED

University of Manchester, Arthur Lewis Building, Manchester, M13 9PL, UK

Tel: +44
-
161
-
275
-
2800/
2804


Email:
cdi@manchester.ac.uk

Web:
http://www.manchester.ac.uk/cdi



View/Download from:

http://www.sed.manchester.ac.uk/idpm/research/publications/wp/di/index.htm


Educators' Guide from
:

http://www.sed.manchester.ac.uk/idpm/research/publications/wp/di/educdi.htm


Table of Contents



A
BSTRACT

................................
................................
................................
......................

1

A. Background and Aims

................................
................................
...............

2

B. Framing and Methodology

................................
................................
........

3

C. Mapping Mobile Phones for ARD: Issues and Evidence

.............................

6

C1.

A
DDRESSING
N
EEDS

................................
................................
................................
...

7

C2.

A
DOPTION AND
U
SE

................................
................................
................................
..

8

C3.

O
UTPUT AND
I
MPACT
................................
................................
...............................

10

D. Conceptual Approaches and Methodological Issues

................................

14

D1.

C
ONCEPTUAL
A
PPROACHES

................................
................................
.......................

14

D2.

M
ETHODOLOGICAL
I
SSUES

................................
................................
........................

16

E. Research Gaps and Future Research Directions

................................
.......

21

E1.

G
APS IN
I
SSUES AND
E
VIDENCE

................................
................................
...................

21

E2.

G
APS IN
C
ONCEPTUAL AND
M
ETHODOLOGICAL
A
PPROACH

................................
..............

23

E3.

C
ONCLUSIONS AND
F
UTURE
R
ESEARCH
A
GENDA

................................
............................

24

F. References

................................
................................
..............................

27



Manchester Centre for Developm
ent Inform
atics Working Paper 50



1

Mobile Phones for Agricultural and Rural
Development:

A Literature Review and Future
Res
earch Directions


Richard Duncombe

Centre for Development Informatics

IDPM, University of Manchester, UK

Email: richard.duncombe@manchester.ac.uk


2012



Abstract


This paper provides a systematic review of the potential and the limitations of mobile
phone
s in the delivery of rural services for agricultural and rural development (ARD) in
developing countries. Evidence from published research is framed according to the design,
uptake, usage and impact of mobile phone
-
based innovations through a critical rev
iew of
the literature encompassing both acad
emic and practitioner sources.


The review indicates a rapid expansion of research in recent years, and a growing number of
primary research studies that have developed rigorous methodologies for data collection
and analysis, with welcome contribution from developing country institutions and
researchers. There is positive linking with mainstream research disciplines and specific
attempts to develop n
ew theoretical understandings.


Gaps in the literature suggest a
reas where future research priorities may lie, relating to
assessment of information and service needs, consideration of indicators of sector
performance and productivity, assessment of broader impacts at the community level,
greater focus on methodologies

that emphasise user involvement,
and
expansion of
qualitative approaches which could form the basis for theorising and greater geographical
diversity. There is also need to investigate new trends that are emerging due to the
convergence and integration o
f new information and communication technologies.


Manchester Centre for Developm
ent Inform
atics Working Paper 50



2

A. Background and Aims


Rural services are at the heart of successful agricultural and rural development (ARD) in
developing countries. Effective delivery of services is defined by Poulton et al (2010:14
13)
as... “essential if small farms in high potential areas are to intensify production, contribute
to economic growth and reduce poverty”. Efficient and effective services are required to
overcome high unit (non
-
labour) transaction costs and to access cap
ital, markets and
technical inputs, thus enabling small producers to compete successfully with larger
counterparts. In recent years, rural services have
diversified from

agricultural extension and
those delivered by government agencies
to include a wider
range of organisational forms


encompassing non
-
governmental organisations (NGOs), public
-
private partnerships, and
increasingly, the private sector. A transformed institutional basis for service delivery has
broadened the scope of intervention


away fr
om solely providing inputs for improving
yields with upgraded production technologies and techniques


to include support for
marketing, market access and quality assurance of produce (Feder, Birner & Anderson,
2011). There is also greater emphasis on how

small
-
holder farmers and other rural
producers can be engaged more effectively in local, regional or global supply chains


an
expanded agenda that has taken on environmental and social concerns, involving new forms
of community mobilisation, organisation
, learning and interaction (Davis, 2008).


Set against this backdrop of rural transformation, the advent of the mobile phone is
stimulating a revolution in rural connectivity for small
-
holder farmers and other small
-
scale
rural producers in developing coun
tries. Infrastructure networks have expanded rapidly
i

and for many rural producers the mobile phone is enhancing communication, information
exchange, and innovation in service delivery (Donner, 2009;
Tickner, 2009;

Parikh et al,
2007). Mobile phone
-
based

services have proliferated in recent years, providing new ways
to access price and market information, and coordinate input/output resources including
transport and logistics, finance and production techniques (Qiang et al, 2011;
Gakuru,
Winters & Stepman
, 2009)
. Successful innovations


that have been scaled


have tended to
be market
-
driven such as Esoko (formerly Tradenet)
ii

in Ghana that provides electronic
means to pull data (concerning market prices, buyers, sellers, inventory, transport, etc) and
pu
sh it back via mobile phone interfaces to thousands of small
-
holder farmers that are
participating in a wide range of produce markets. Personal use of the mobile phone has also
enabled rural producers to interact directly with end
-
user markets, traders, s
uppliers,
extension services and with each other.


The aim of this review is to analyse and understand the potential and the limitations of
mobile phones in the delivery of rural services for agricultural and rural development (ARD)
in developing countries
. This will be achieved by assessing the available evidence
concerning the design, uptake, usage and impact of mobile phone
-
based innovations
Manchester Centre for Developm
ent Inform
atics Working Paper 50



3

through a critical review of the literature encompassing both academic and practitioner
sources. This review wil
l...




Provide analysis of literature and synthesis of research concerning mobile phones for
ARD in developing countries


with reference to mobile services that impact upon
small
-
holder farmers and other small
-
scale rural producers.



Categorise and analyse
conceptual approaches for understanding mobile phones for
ARD in developing countries.



Assess methodologies used to carry out research studies that have been conducted
into mobile phones for ARD, and evaluate the evidence from those studies.



Identify key r
esearch trends and gaps relating to concepts and methodologies,
evidence presented, issues addressed and questions raised, and provide an agenda
for future research.


A review of evidence in this area is opportune given growing global interest in the role
of
mobile phones for development (M4D) from donors, NGOs, governments and the
commercial sector. Expanded rural connectivity, and in particular the growth of mobile
services, appears to be stimulating a new wave of optimism concerning the potential of
mob
ile applications (and ICT more generally) to deliver greater scalability, replication and
sustainability than has been achieved in previous phases of ICT for development (ICT4D).
iii


B. Framing and Methodology


In framing the research area, there is need to
move beyond the type of taxonomies that
predominate in the predominantly practitioner
-
orientated literature. These are illustrated
by
Vodafone (2011) that suggests mobile
-
based ‘solutions’ that can provide dedicated
systems for mobile payments, farmers he
lp
-
lines, produce traceability and tracking systems,
agricultural trading platforms, etc. IDRC (2008) simply list the technologies, ranging from
Geographical Information Systems (GIS) using mobile mapping and Personal Digital
Assistants (PDAs) to mobile c
ellular, Internet and Web
-
based applications.
Whilst useful for
providing compendiums of existing initiatives, these approaches to framing mobile phones
in ARD have a key drawback. Although they pay attention to prescribing broad areas of
application, th
ey are technology
-
led


viewing the technologies as readily available and to
be picked ‘off the shelf’ rather than part of more complex and dynamic processes of rural
development and transformation in service delivery. This tends to give rise to insuffici
ent
understanding of the diverse agricultural and rural contexts within which mobile
applications are being introduced.


To address this shortfall, this review will be framed within a 2
-
dimensional matrix that
incorporates a temporal lifecycle for mobile p
hone innovation (based on Heeks, 2007) and a
further distinction according to the methodological basis for the study. Additionally, a
Manchester Centre for Developm
ent Inform
atics Working Paper 50



4

categorisation is developed in Section 4 for emergent theoretical approaches based on a
taxonomy used by Gregor (2006).


The temporal lifecycle identifies the positioning of the identified study according to four
interconnected parameters.



Design:

assessment of the requirements for mobile phone innovation that includes
the diversified information and service delivery needs o
f potential users.



Adoption:

an understanding the factors that either drive or constrain uptake, and
patterns of usage of mobile phones for ARD.



Output
: the identification and measurement of the tangible costs and benefits of use
of mobile phones such as i
n the provision of information or transactions.



Impact
: the evaluation of broader productivity or welfare gains or threats that result
from the application of mobile phones for ARD.


The lifecycle model allows for a broad range of research to be considered

in the review that
draws upon divergent theoretical and methodological perspectives


ranging from
practitioner
-
led action research that has evolved out of the design processes attached to
specific applications to more academically
-
orientated impact studi
es that analyse larger and
more representative samples. This broad spectrum of research incorporates both a
provider and user perspective. Thus, on the one hand, research concerning
active
innovation

in mobile
-
based service delivery is surveyed


studies

focused on analysing
specific initiatives that are designed to incorporate the use of mobile phones into existing
rural service interventions. On the other hand, it is also important to consider how access
to rural services results from
passive diffusion

of mobile phones


to understand the
patterns of usage and the degree of uptake of services that is stimulated by pre
-
existing
ownership of mobile phones and access to networks.


Methodological approach was selected as a secondary axis for coding of the a
rticles
primarily due to ease of identification, and was
distinguished in a fourfold manner: a)
quantitative studies which are more representative in terms of sampling, but possibly
contribute less to theoretical understanding. These comprise
12 research

articles all of
which rely on field experiments and survey instruments (using structured/semi
-
structured
questionnaires) or the analysis of secondary data;
b) qualitative studies which largely focus
on analysis of individual case studies, which make no cl
aims to the general population, but
tend to provide more in depth data concerning processes and contribute more to
explanatory theory. These comprise
10 research articles


seven of which build case
studies.

Also added were: c) mixed methodology studies
which totalled
18 research articles


17 of which relied on survey instruments (using interviews or semi
-
structured
questionnaires); finally, d)

those containing a strong action
-
participative element.



comprising 3 research articles
.

The spectrum of meth
odological approaches is broad and in
many cases there is crossover of data collection methods and data analysis techniques. For
Manchester Centre for Developm
ent Inform
atics Working Paper 50



5

example, some mixed methodology studies tend more towards quantitative analysis, whilst
some purely quantitative studies make
use of participatory methods to formulate and test
suitable research instruments.

The outcome of the coding of the articles is represented
according to the distribution of articles in Fig 1 (overleaf), and analysed in more detail in
Section 4 of the review
.
iv


The review identifies published research that is peer reviewed from academic journals,
conference and working papers. However, because many key developments in the field are
practitioner
-
led, some non
-
peer reviewed sources are also surveyed and analys
ed including
consultant and official reports and other occasional and published papers.
v

In line with
Table 1, the review is time limited


encompassing research published during the past 10
years with a cut
-
off point of 31
st

March 2012. Given the cross
-
disciplinary nature of the
topic, research is drawn from a wide range of academic disciplines which include
development studies, economics, technology and innovation, information systems (IS) and
information & communication technology for development (ICT4
D). The scope of the
review includes those studies assessing the application of mobile technologies for ARD in
developing countries and English language sources only.
The review includes research
articles that report the results of analysis of primary da
ta, with emphasis on
critical
assessment of findings according to the theoretical and methodological approaches
followed. Also included are studies that analyse data from secondary sources, either from
official sources such as provided by international or
ganisations (e.g., the International
Telecommunications Union


ITU) or through analysing data extracted from pre
-
existing
associated research (such as household panel data drawn from surveys conducted in rural
areas).


Table 1. M4ARD Research Timeline an
d Country Focus


Final year
of data
collection

No of
research
articles

Country focus

2000

1

Botswana

2001

1

Bangladesh

2002

1

India

2003

-


2004

2

India, Mozambique, Tanzania, RSA

2005

3

Uganda, Tanzania, Ghana

2006

8

India, Rwanda, Kenya, Nigeria,
Tanzania, Niger, Philippines

2007

2

Uganda, Ghana

2008

7

India, Uganda. Tanzania,

2009

8

Ghana, Uganda, Malawi, Kenya, Sri Lanka, India

2010

9

Ghana, Uganda, Malawi, Kenya, Sri Lanka, India, Bangladesh

2011

1

Tanzania

Total

43



Manchester Centre for Developm
ent Inform
atics Working Paper 50



6

C. Mapping Mobile P
hones for ARD: Issues and Evidence


As a first step, the overall distribution of the articles was gauged according to the 2
-
dimensional modelling outlined (temporal lifecycle and the methodology employed).
Additionally, it is indicated whether the evidenc
e presented is based on analysis of primary
or secondary data (Fig 1).

Fig 1. Mapping Mobile Phones for ARD Research


Studies assessing
information and
communication
needs

Studies concerned
adoption

processes and
patterns of
use

Studies concerned
with ass
essing
outputs

Studies concerned
with broader

impact

on welfare

Quantitative










Mixed
methods










Qualitative









Participatory
/action











Normal: analysis of data collected from primary sources (38 articles)


Ita
lic
: analysis

of data collected from secondary sources (5 articles)


Refer to list of references for article number


The following sections will discuss research findings for mobile phones for ARD in terms of
the evidence presented and the issues and quest
ions raised in the articles. The discussion is
structured according to the temporal lifecycle model, also recognising where areas of
overlap exist (as suggested in Fig 1).


25

10, 11

9,
20
,
22

2
, 17,
24
,
31
,
40,
41

18,

13
19
28

6,

32


12, 14, 15,
21, 26, 39


1,

3


7, 8, 33

36

5


16, 29, 30


37,
38
, 43


4, 35


23, 42


27


34


Manchester Centre for Developm
ent Inform
atics Working Paper 50



7

C1. Addressing Needs

The number of studies that analysed mobile phones for ARD exc
lusively in terms of the
needs of farmers and the rural community were few. Prominent was De Silva &
Ratnadiwakara (2010) who randomly sampled 300 farming households in Sri Lanka across
four traded vegetable markets, and attempted to understand informatio
n search costs for
core operations along the agricultural
-
farmer value chain. They found that 70 percent of all
transaction costs related to information search costs (the transaction costs themselves were
recorded as 15 percent of the total costs incurred
). Costs were accounted for both in direct
financial expenditure and the opportunity costs of time expended. Similarly, Furuholt &
Matotay (2011) view the mobile phone as a means whereby the high transaction costs
associated with information market failu
res and inefficiencies in the business environment
can be reduced, thus overcoming costs incurred in the coordination of economic activities
related to: a) accessing inputs (infrastructure; production technology, knowledge, finance,
materials, learning/tra
ining, etc) and, b) reaching output markets (either directly or through
market intermediaries) as well as monitoring financial transactions and consulting with
experts. Typically, the studies surveyed view search costs and the asymmetric relationships
tha
t govern price setting as the most significant informational impediments producers face
(e.g., Islam & Gronlund, 2011; Martin & Abbott, 2011). Such high costs normally add to the
market price of products and affect competitiveness in the market.


Ndiwalan
a et al (2010) identify information gaps outside of the economic sphere that are
equally, if not more critical, for farming households. This expands the definition of needs
into a more diversified set of livelihood concerns that are important for overcomi
ng the
broader social, political, location and environmental
-
climatic constraints that rural
producers face. Kameswari et al (2011) surveyed 132 farmers across 8 villages in 4 districts
of the Indian Himalayan region and found wide variations between dist
ricts (and villages) in
terms of crops grown, scale of production, water supply, types of soil, etc. This gives rise to
diverse vulnerability contexts and differing needs for information
-
related services often
between districts and settlements in quite c
lose proximity, and which are more or less
active in information seeking. A similar perspective was taken by Masuki et al (2010)
highlighting how differences in the cultural and social make up of different parishes within
the same district of Uganda gave
rise to different needs for information, as well capabilities
to make use of information. These livelihoods perspectives support the view that fostering
sustainable rural production involves addressing a wide range of interconnected constraints
which may
be longstanding and entrenched within the realities of rural life, and reach into
broad and diverse development concerns of environmental protection and conservation,
gender imbalances, political participation, health and education (Feder, Birner & Anderso
n,
2011).


Studies emphasise that information received (either as voice or data) needs to be made
usable and actionable, and this requires further information chain resources (Duncombe &
Manchester Centre for Developm
ent Inform
atics Working Paper 50



8

Heeks, 2002) which are both intrinsic


pre
-
existing knowledge and th
e necessary
capabilities to act


and overt


the necessary infrastructure, technical and specialist
expertise that are required to make effective use of information received. In research from
Uganda, Burrell & Matovu (2008:3) concur, suggesting that the.
.. “main challenge to
providing information services to rural Ugandans rests on providing dependable accurate
information that is ‘actionable’. Whether information is actionable depends especially upon
what capital and what social connections are required

to make use of it”. Information may
also be viewed differently by recipients than by providers. For example, recipients may not
view information as a distinct resource, but as broader advice, the veracity and utility of
which, is strongly associated wit
h the source, and most commonly demonstrated through
human interaction (Burrell & Matovu, 2008). In this regard, mobile phones are strongly
linked in the literature


and in the eyes of users


to maintaining and improving social
networks


particularly f
amily and personal ties


due to the greater scope of
communication they afford (Sife et al, 2010; Goodman, 2007; Donner, 2007; Souter et al,
2007).


Historically, the complex information needs of rural producers have been pursued through
these personal an
d social networks, and mediated through face
-
to
-
face contact. Traditional
networks of communication tend to be better aligned with the interests of rural dwellers
and information sources may embody a certain level of trust (Molony, 2007). Indian
research
(Islam & Gronlund, 2011) and research from Kenya (Okello et al, 2010) and Uganda
(Martin & Abbott, 2011; Kashem, 2010) found that input suppliers, fellow traders and clients
were the main sources of market information, thus reinforcing the view that pre
-
ex
isting
informal networks of communication represent the bedrock of rural information systems
(Duncombe & Heeks, 2002). An interesting perspective was provided by Fu & Akter (2011)
who investigated how mobile phones impacted upon pre
-
existing extension net
works in
India where phones were carried and used by
munnas


assistants to agricultural specialists
travelling between villages. The study takes a randomised population (treatment and
control group) and measures proxies for adoption, knowledge generation
, attitude and
awareness amongst a sample of 698 farmers (where there existed a 50% illiteracy rate)
measuring before and after intervention differences. Evidence suggests that the amount
and quality of services and the speed of delivery had been improved

significantly as a result
of the mobile phone based intervention with quality of services measured as 74 percent
higher than what was available before the ICT (mobile) enhanced services were introduced.
Also, disadvantaged farmers


who were starting from

a lower base of awareness and
knowledge


made greater use of the intervention that those that were better off.


C2. Adoption and Use

The most prominent early mobile phone
-
based intervention was the Grameen village phone
operator (VPO) programme in Bangla
desh. The programme provided important
empowerment opportunities for women, but the detailed study of Aminuzzaman et al
Manchester Centre for Developm
ent Inform
atics Working Paper 50



9

(2003) found that both phone operators (predominantly female) and users of the services
offered (predominantly male) were more likely t
o be traders and entrepreneurs than
farmers, and only 4 percent of users were illiterate (compared with an 80 illiteracy rate
nationally at the time). More recent studies still identify early adopters and more intensive
users as better off with a high cor
relation between mobile ownership, use of a bank account
and greater asset endowment (e.g., Katengeza et al, 2011). Intangible characteristics of
farmers (such as confidence and aspiration) are also highly correlated with more intensive
use, and the const
raints of ownership and use are exacerbated for the less well off and the
poor. These disparities were evident in the findings from a number of studies from different
countries and regions represented in the literature. However, some contrasting findings

were also evident. Martin & Abbott (2011) suggest that mobile phones are having a
democratising (levelling) effect with no significant difference in ownership between leaders
and non
-
leaders identified in rural communities. Another study from a poor remo
te area
with difficult terrain shows relatively high adoption rates by poor farmers such as an 80
percent ownership rate within the sample reported by Kameswari et al (2011) for the Indian
Himalayan region. Similarly, two empirical studies based on a repr
esentative sample of
phone users demonstrated that greater distance of farmers from markets (i.e., greater
remoteness) implied greater intensity of phone use (Katengeza, Okello & Jambo, 2011;
Lwasa et al, 2011).


A further set of issues arose in relation t
o predisposition towards individualised or group
-
based ownership and use. This is pertinent, given that in developing countries rural
producers look to mitigate risk by acting collectively through various forms of community
-
based organisation. Typically
this takes the form of producer groups organised around the
cultivation and marketing of specific crops or produce. Extension support also tends to be
group
-
directed such as training and visit (T&V), farmer field schools (FFS) and more
contemporary market
-
based fee
-
for
-
service aimed at farmers groups, cooperatives or whole
communities (Davis, 2008). Although the literature highlights a number of cases of
innovation in mobile
-
based services built upon group
-
based participation (
Masuki et al,
2010;
Gandi et

al, 2009;
Kithuka et al, 2007; Veeraraghavan et al, 2009)
, there is also
evidence that ownership and use of mobile phones is allied to an emerging individualised
culture. Research conducted in Uganda by Masuki et al (2010) indicated relatively low levels

of usage of communally provided phones (available to women’s groups via village
information centres) compared with their use by individual farmers (predominantly men)
which was expanding rapidly amongst lower income groups. In associated research from
Ug
anda, Lwasa et al (2011) found that owners of phones are less inclined to participate in,
and less aware of, group
-
based initiatives for information sharing and delivery, whereas
existing group members are more aware and more inclined to participate. Howe
ver, studies
showed variation in levels of awareness and usage of services that could potentially be
accessed via mobile phones. Most commonly, this was due to poor technological skills and
Manchester Centre for Developm
ent Inform
atics Working Paper 50



10

experience (of the farmers that could potentially access the serv
ice) and in many cases that
the information provided lacked relevance to their needs.


Another common finding was that use of voice predominates (e.g., Furuholt & Matotay,
2011) with significant constraints on using other formats such as SMS with only 5 pe
rcent of
users recorded as regular text users by Islam & Gonlund (2011). Similarly, Okello et al
(2010) found that 79 percent of rural market traders used mobile phones in their trading
activities, with 77 percent preferring to use voice rather than 2 per
cent that used SMS. This
was because traders preferred a system that allowed interaction with the broad range of
participants in any given produce value chain including agents, brokers, assemblers,
wholesalers and final purchasers


and voice best facilit
ates this. The preference for voice is
also illustrated by

Veeraraghavan et al (2009) showing how Indian farmers benefited from a
move away from PC
-
Internet
-
data
-
based systems located in rural kiosks (which showed a
high failure rate) to a mobile network
voice
-
based system that was found to be less
vulnerable, easier to maintain, more accessible to farmers in remote areas, and delivered
more up
-
to
-
date provision of information and notification of prices.
vi


C3. Output and Impact

A distinction has been drawn

in this review between
passive diffusion

of mobile phones into
communities (largely market driven) and
active innovation

in mobile
-
based service delivery
(largely intervention/donor driven). Theories of reasoned action suggest that passive
diffusion is s
timulated by the perceived benefits of ownership (Venkatesh et al, 2003) and
much of the evidence from the studies surveyed supports this proposition. Of particular
note is the study of Jensen (2007) that carried out a longitudinal analysis of 300 sardine

fishing units (boats) operating off the coast of Kerala in India. The findings were almost
wholly positive with the use of mobile phones strongly linked to reduced price dispersion
across markets for landed fish and an almost complete elimination of fish

wastage. Whilst
wealthier fishermen (early adopters) were the greatest beneficiaries, small
-
scale fishermen
(including those without mobile phones) also benefited due to the spill
-
over effects of more
efficient markets. Abraham (2007) provides less empi
rical rigor, but qualitative results
largely confirm the positive outcomes associated with Jensen. Interestingly, however,
respondents viewed the use of mobile phones as less important for their livelihoods than
the introduction of improved production (fi
shing) technology and improved roads and
transport. Other studies of use of mobile phones by fishing communities largely confirm the
results from Kerala including those of Salia et al (2011) where 22 percent of the sample used
phones to obtain market pric
e information from multiple landing sites which led to better
market coordination and market prices.


In the agricultural sector, Egyir et al (2011) identified mobile phones as the pre
-
dominant
communication technology amongst farmers in Ghana, measuring a
n increased speed of
price transmission in maize markets


but no marked decrease in transaction costs in
Manchester Centre for Developm
ent Inform
atics Working Paper 50



11

markets where exchanges were dominated by traders that lack literacy and rely heavily on
visual inspection. Whilst 80 percent of traders and 48 perce
nt of farmers surveyed used
mobile phones to access information, traditional means of collecting and exchanging
information had not changed (such as travelling to the market) but use of phones speeded
up pre
-
existing processes. For mobile users, this led
to the trading of larger volumes, better
prices and slightly larger margins


but only marginal increase in transaction costs (due to an
increase in the net cost due to the costs of mobile ownership and use). Similarly, Muto &
Yamano (2009) from analysis
of data collected between 2003 and 2005 in Uganda found
that mobile network expansion (from 41 to 87 percent coverage) had positive effects on
market participation. The effects were found to be more beneficial for farmers in remote
areas and particularly
for those producing perishable crops such as bananas. Similarly, Aker
(2008) investigated the impact of cell phones on grain markets in Niger identifying positive
arbitrage (reduced grain price dispersion and variations across markets) resulting from a
re
duction in search costs and hence transaction costs, as well as lower grain prices (3.5%
reduction from 2001 to 2006). In common with Muto & Yamano, phone use was found to
have greater impact when travel costs were higher


for markets that were more remo
tely
located and unconnected by paved roads (Burrel & Matovu, 2008; Kithuka et al, 2007).
Phones also caused traders to change their behaviour


with a greater number of markets
searched and more contacts and sales in more markets. These studies also ide
ntify a
‘network effect’ meaning that cell phones have higher impact (on price dispersion, for
example) once more markets are covered by the network


with Acker (2008) suggesting
diminishing returns over and above 75 percent network coverage.


Data and ev
idence concerning active innovation in services is less apparent in the review.
Notable recent studies include that of Subervie (2011) evaluating the impact of SMS
-
based
alerts for farmers via
Esoko

where econometric modelling of spatial arbitrage conditi
ons
found a significant effect on prices with a 10 percent increase amongst the treatment group
of 500 farmers to whom mobile phones were distributed in the northern region of Ghana.
In contrast, Fafchamps & Minten (2011) gauged the benefits that Indian f
armers derive from
market and weather information delivered to their mobile phones via a commercial service
called Reuters Market Light (RML). A robust estimation technique was used to generate
findings for treatment and non
-
treatment groups comprising 93
3 farmers across 100 villages
(20 in each of 5 villages) in the Maharashtra region. There was some evidence that use of
RML positively impacted upon spatial arbitrage and crop grading, although the effect was
small. However, no significant effect was mea
sured on the price received by farmers for the
produce, nor on crop added value, crop losses resulting from bad weather or the likelihood
of changing crop varieties and cultivation practices. In this case, better price information
did not result in higher

prices paid to farmers and this is explained due to the lack of
alternative market destinations and the lack of opportunities for arbitraging by farmers.
Overall, the study found a small number of clients for the service in aggregate across the
study are
a and stagnation in take up of the service over the study period. Also in India,
Manchester Centre for Developm
ent Inform
atics Working Paper 50



12

Mittal et al (2010) surveyed the use of IFFCO
-
IKSL, Fisher’s Friend (and RML), each of which
provide subscription
-
based messaging services for packaged information concerning

weather, crop advisory tips, market prices, input availability and government schemes.
Increasing numbers of subscriptions had brought some benefits to some famers, but
constraints on these ‘stand
-
alone’ services included lack of awareness of their exist
ence and
what they can offer, lack of customisation and updating of content, concerns over timeliness
and reliability of information and lack of use of local languages.


More positively, Katengeza et al (2010) demonstrated benefits accruing to farmers from

improvements in spatial rice market integration comparing a period before and after the
introduction of the Malawi Agricultural Commodity Exchange (MACE) (involving use of
mobile phones for the communication of market information) across nine selected mar
kets.
The post MACE period saw improvements in market integration (and arbitraging) enabling
farmers to access bigger markets and get better prices. However, key intervening factors
(that caused high variation in results) were the distance of markets from

other markets and
poor transport links, which impeded trade. In a separate study making use of a sample of
participants and non
-
participants in MACE, Katengeza et al (2011) found that greater
distance from markets was correlated with greater intensity of

phone use, which supports
findings set out previously. Largely positive results were also reported by Annamalai & Rao
(2003:19) in an evaluation of e
-
Choupal


a pioneering Indian initiative that makes use of
mobile phones to connect farmers into the agr
icultural supply chains of ITC


a large Indian
agro
-
business. In this case... “farmers benefited from more accurate weighing, faster
processing time, and prompt payment, and from access to a wide range of information,
including accurate market price know
ledge, and market trends, which help them decide
when, where, and at what price to sell.” The research suggested that farmers received an
average of 2.5 percent higher income (though it is unclear if this is due to higher price or
lower transaction costs)

as well as lower prices for inputs and other goods, higher yields,
and a greater sense of empowerment. The losers from the new value chain model were
commission agents, labourers at the non
-
ICT
-
enabled government markets, and shops near
those markets, as

well as women who had no access to the system at that time.


The preceding studies demonstrate primary interest in impacts on farmers and the markets
they trade into. Studies also highlight the impact of phone use on transactional
relationships within va
lue chains


and in particular the changing role of intermediaries or
traders. Overå (2006) found that both producers and traders benefited considerably from
the use of mobile phones after their introduction in 2001 in Ghana. Speed of
communication allow
ed for more efficient information flows within the network of value
chain actors, which in turn, saved time and reduced transportation costs. This led to better
matching of supply and demand, and improved monitoring of compliance within the terms
of tradin
g contracts. One effect was that early adopters of mobile phones strengthened
their existing trader relationships and networks, which were built on strong lineage
-
based
Manchester Centre for Developm
ent Inform
atics Working Paper 50



13

social structures. New market entrants managed (through using mobile phones) to quick
ly
cement good trading reputations and facilitate the building of more efficient trading
networks. By contrast, existing traders and new entrants without phones were not able to
attain these advantages, although it is not clear whether they were financiall
y worse off as a
result.


In contrast, Jagun et al (2008) found little change to the structural characteristics of
transactional relationships which remained localised and intermediated. In fact the use of
mobiles had consolidated the power and influence o
f market intermediaries (middlemen)
due to their role in accessing complementary market resources such as access to capital and
materials. Similarly, a study carried out in India points towards the strong position of
established commission agents and trad
ers in local supply chains who are the major price
setters (Mittal et al, 2010). Traders and input dealers also provide an important source of
information


particularly related to agricultural technology and techniques. One of the key
benefits identifie
d for phone use was as a basic communication device enabling farmers to
communicate more effectively within their pre
-
existing networks


ranging from 10
-
30 calls
per day


thus reinforcing those networks. Finally, Molony (2007) concludes that trust in
ec
onomic relationships does not exist independently of social ties and these tend to be
embedded in personal interaction which necessitates face
-
to
-
face communication, whilst
Masuki et al (2010) found that the complex nature of interactions between service p
roviders
and farmers largely mitigated against use of phones.


Manchester Centre for Developm
ent Inform
atics Working Paper 50



14

D. Conceptual Approaches and Methodological Issues


D1. Conceptual Approaches

A picture of the conceptual approaches used to research mobile phones for ARD in
developing countries can be gaine
d from Table 2 where the research articles are identified
together with key references to their antecedents


previous research cited in the article
upon which the conceptual approach was based.


Table 2. Mapping Conceptual Approaches

Lifecycle

Conceptual
approaches identified

Classifi
cation

Antecedents cited

Article

Needs/

Design







Adoption/

Use









Output









Impact

Information needs modelling

C

Heeks (2009)

Duncombe & Heeks
(2002)

32, 28, 8

Awareness/knowledge assimilation

C

-

11

Value c
hain/supply analysis

C

Porter & Millar
(1985)

4, 6, 34, 12, 16, 36

Technology acceptance model (TAM)

B

Davis (1989)

25

Capital endowments; incentives

C

-

21

Diffusion theory

B

Rogers (2003)

26

Modelling social capital formation
Social/economic netw
ork analysis

B

Granovetter (1973)

Andrade &
Urquhart (2009)
Barr (2002)

Coleman (1988)

7, 14, 15, 35

institutional economics; asymmetrical
information; trust in economic
relationships

B

Fafchamps (2004)

3, 24, 29, 30

Micro
-
economic modelling


transact
ion costs; information search
costs

A

Doward et al (2003)

Barrett (2008)

1, 33

Micro
-
economic modelling


changes
in prices; spatial arbitrage/market
integration

A

Fafchamps & Hill
(2005)

Stigler (1961)

2, 9, 10, 17, 20, 24,
31, 40; 41

Livelihoods anal
ysis/technology
appropriation at base of pyramid


C

Ellis (2000)

Orlikowski (1992)

5, 37, 38, 39

Note: Eight studies are not included as they indicated no significant theoretical contribution


In line with the distinction drawn by Gregor (2006) the first
category (A) identifies theories
that are applied or tested to predict how mobile phones bring about changes in key
Manchester Centre for Developm
ent Inform
atics Working Paper 50



15

variables such as transaction costs and market conditions (e.g., prices paid and arbitrage
across markets). These studies emphasise quantit
ative analysis. In this vein, Labonne &
Chase (2009) and Katengeza et al (2010) measure the effects of mobile phone use on
information asymmetry


proposing that better access to information (via phones) allows
farmers to strike better price deals within
their existing trading relationships, and to make
better choices about where to sell their produce. Jensen (2007), Acker (2008) and Egyir et al
(2011) move beyond measuring narrow price effects and look at differential prices
(dispersion or arbitrage acro
ss markets for the same produce). Here, better information is
hypothesised to lead to a better aggregated market performance and positive welfare
outcomes for producers and consumers (as well as users and non
-
users of mobile phones).
The hypothesis of Mu
to & Yamano (2009) is narrower, suggesting that expansion of mobile
phone coverage induces greater market participation of farmers that produce perishable
crops in remote areas. Fafchamps & Minten (2011), Subervie (2011) and Svensson &
Yanagizawa, (2009)
also focus more narrowly on the application of quantitative estimation
techniques to evaluate the impact of specific interventions identifying key constructs
concerning spatial arbitrage, farm gate prices and market efficiency indicators such as crop
value

added and crop losses, whilst Fu & Akter (2011) seek to quantify and measure proxies
for intangible impacts (e.g., enhanced knowledge and awareness of farmers).


A second set of theoretical approaches (B) can be grouped as explanatory (rather than
narrowl
y predictive) and they are located in diverse disciplines (anthropology;
information/management studies; development studies). These include both deeper
theory, and framework
-
based approaches derived from a body of theoretical work.
Explanatory theories
try to understand the often complex causal relationships between
variables, and the resultant outcomes and impacts for ARD within a community. These
approaches are most often framed within existing bodies of theory and can employ
quantitative, qualitative

or mixed methodologies. Most notably, Overa (2006) following the
work of Granovetter (1985) argues that adoption and use of mobile phones enhances trust
building in trading networks, thus facilitating a higher number of transactions in uncertain
environm
ents where trust is at a premium. Donner (2007) suggests a strengthening of pre
-
existing relationships within networks arising from phone use


phones are used to amplify
and strengthen existing trading relationships which are often family and lineage bas
ed. The
primacy of pre
-
existing channels for information exchange and the overriding importance of
personal face
-
to
-
face communication is theorised by Jagun et al (2008) and Molony (2007).
In line with Donner, both Islam & Gronlund (2011) and Goodman (20
05) conceptualise
mobile phones as users see them


as primarily social rather than economic tools. Lwasa et
al (2011) takes a different approach, making use of innovation theories linked to the
technology acceptance model (TAM) to gauge how socio
-
cultura
l factors (relating to farmer,
household, farm, assets and location specific variables) condition adoption and use of
mobile phones in agricultural transactions.


Manchester Centre for Developm
ent Inform
atics Working Paper 50



16

A third set of approaches (C) use frameworks, models (such as business models), single
concep
ts or categories that are applied but which are not strongly grounded in theory.
Most prominent are the producer value chain (Furuholt & Matotay, 2011; De Silva &
Ratnadiwakara, 2010; Annamalai & Rao, 2003) where information search and other
transaction c
osts are modelled according phases of pre and post harvest activity


differentiating the primary value chain according to land preparation, seed purchasing,
growing, harvesting and selling. Other studies take a broader view modelling sector value
chains
(Salia et al, 2011; Okello et al, 2010) that include not only primary activities but also
support activities and extended sector value chain (or supply chain) comprising actors
located upstream (supplying farm inputs) and downstream (responsible for possib
le further
processing, marketing and distribution of produce) as well as final consumers. Also
prominent is the modelling of information needs (Mittal et al, 2010; Ndiwalana et al, 2010;

Duncombe & Heeks, 2002) which are seen to vary according to market re
ach


into local,
regional and global markets. Producers serving local markets are reliant on information
delivered informally through local networks of communication, where trust and risk
reduction are major factors that govern their dependence on those
networks. Timeliness of
information is a serious failure of the information delivery system currently used, and a
significant aspect of their vulnerability to changes in the surrounding environment. A higher
degree of integration of producers into market
systems


extending to regional or global
value chains


demands an increased volume and complexity of information as the value of
information (and use of ICT) is better recognized. Producers that lack this capacity and
opportunity are likely to become ma
rginalized and excluded


particularly from global value
chains. Another prominent group of studies look at how use of mobile phones in market
settings can work better for the poor, addressing not only information (market) failures but
also livelihood obje
ctives


to build and acquire assets and reduce vulnerability. Thus in
addition to providing a means to transact goods and services, mobile phones may also have
potential to enhance the empowerment, opportunity and security of the poor


taking
livelihood
s perspective (Sey, 2011; Sife et al, 2010; Burrel & Mato
vu, 2008; Souter et al,
2007).


D2. Methodological Issues

Research strategies followed in the studies encompass use of field experiments, field studies
(especially case studies or ethnography), actio
n/participatory research and compilation and
analysis of data from secondary sources (such as existing surveys or panel data). The review
evaluates the robustness of the research methods by assessing: 1) means of sampling and
sample size; 2) the setting o
f baselines or provision for counterfactual explanations; 3)
recourse to validity testing or discussion of validity concerns; 4) whether research is cross
-
sectional (snapshot) or longitudinal (giving time
-
span); 5) the extent of method guidance
that would
be sufficient to replicate the research. Finally, it is indicated whether the article
is peer reviewed (which most are).


Manchester Centre for Developm
ent Inform
atics Working Paper 50



17

The largest group of studies falls between the two methodological poles employing mixed
methodology approaches most commonly combinin
g a questionnaire or interview survey
with more in
-
depth key informant interviews or focus groups. With the exception of
Aminuzzaman et al (2003) mixed methodology approaches are snap
-
shot surveys that
employ varying degrees of cross
-
sectional design. In

terms of contributing to stated
programme and policy objectives they achieve their goal, but the rigour and transparency of
their stated methods varies. Mixed method studies are less clear about whether or how
they address the counterfactual, and tend no
t to employ sufficient triangulation of data
sources to demonstrate validity. They tend to use anecdotal examples and measure
perceptions of users of stated variables, rather than the variables themselves (e.g., Martin &
Abbott, 2011). There is also limi
ted method guidance. However, mixed method studies that
are effectively designed and properly conducted provide useful exploratory findings.

Manchester Centre for Developm
ent Inform
atics Working Paper 50



18

Table 3. Mapping Methodological Approaches

No

Tax

Study focus

Data type/
collection


Data/sample



Baseline/

coun
terfactual


Validity/

testing

Timing


Method
guidance

Peer
review





Primary data

Secondary data






1

A

-
Economic impact of mobile
phones on the fishing industry in
India

Mixed
methods

-
focus groups

-
survey

-
12 locations

-
172 respondents

None

None

Sam
ple survey cross
checked with focus
groups

Snapshot

Little

PR

2

A

-
Impact of cell phones on grain
markets in Niger


Quant
itative


-
395 traders

-
205 farmers

-
35 markets

Monthly grain
prices over 10
years (1996
-
2006)
across 42 markets


Quas
i
-
experimental


pooled and separate
treatment groups

Alternative explanations
tested by controlling for
confounding variables,
selection bias and
heterogeneity

Multiple time
periods

Detailed

PR (int)

9

A

-
The effect of ICT
-
based market
information services on the
perform
ance of agricultural
markets in Ghana,

Quantitative

-
structured
questionnaire

-
140 farmers


-
11 markets

-
486 traders

-
KI interviews

-
rural focus groups

None

Users and non
-
users
surveyed

Testing for co
-
integration
between source and
destination markets

Snap
shot April
2010

Detailed

Non
-
Peer
Reviewed

PR

10

A

-
Impact of SMS
-
based
agricultural information on India
farmers

Quantitative

-
interviews

-
100 villages

-
20 in each of 5
districts

-
933 farmers
interviewed and
followed up

None

Randomised control
trial

Trea
tment and control
groups

Random sampling

Robust testing strategy

Snapshot

Detailed

PR (int)

11

A

-
The impact of ICT on agricultural
extension services delivery in
India

Quantitative
Randomized
survey data

-
interviews

-
698 farmers
interviewed

-
507 in contr
ol
group


None

Careful identification of
control and treatment
group

Double difference
strategy controls for
farmer, household and
village characteristics

Changes before
and after the
intervention

Detailed

PR (int)

17

A

-
Information, technology, and
marke
t performance, and
welfare in the South Indian
fisheries sector

Quantitative


300 fishing units

None

Baseline established
prior to trial

Random sampling

Robust empirical
strategy

Longitudinal
1996
-
2001


Detailed

PR

20

A

-
The role of ICT
-
based market
infor
mation services in spatial
food market integration in
Malawi


Quantitative


None

Official data from 9
selected markets
across Malawi


Baseline set according
to study period

Causality tested through
econometric means

Longitudinal

1994
-
2004
-
2007

Detailed

PR

24

A

-
Impact of mobile phones on
farmers’ welfare in the
Philippines


Quant
itative


None

Mobile phone
coverage data for
2,400 HH

combined with HH
Baseline established
prior to study

Reliability tests
conducted via estimation
strateg
y

-
Change
measured
between 2 data
collection points
Detailed

PR (int)

Manchester Centre for Developm
ent Inform
atics Working Paper 50



19

data from 135
villages

2003


2006

31

A

-
The impact of mobile phone
coverage expansion on market
participation in Uganda

Quantitative

None

856HH

94 communities

Baseline established

Robust estimation
strategy

2 data points
2003 and 2005

Detailed

PR

33

A

Use of information and
communication tools and
services by rural grain traders in
Kenya

Mixed
methods

-
survey

204 traders

141 in KACE
project area

63 outside area


Users and non
-
users
surveyed

Random sample

Non
e

Snapshot (May
-
July 2010)

Little

PR

40

A

-
Evaluation of the impact of a
Ghanaian mobile
-
based MIS

Quantitative

-
interviews

196 users

203 future users

200 non
-
users

None

Quas
i
-
experimental

Users and non
-
users
surveyed

Preliminary data


bias in
sample

Bef
ore and after
treatment
groups

Little
available as
yet

NPR

41

A

-
The impact of a market
information service in Uganda

Quantitative

-
interviews

Key informant
interviews

Secondary data on
farm gate prices
from UNHS (2005)

Treatment and non
-
treatment groups
surveyed


Robustness checks
employed

Lack of reliability
checking of source data

Before and
after/selection
controls

Some

PR

3

B

-
A study of the village pay phone
of the Grameen Bank in rural
Bangladesh

Mixed
methods

-
survey

-
350 VPOs

-
20 locations

-
158 u
sers

-
85 operators

-
55 key informants


None

Included 50 non
-
users
from control area

Triangulation between
different data sources

Data collected
1999
-
2000

Little

PR

7

B

-
The use of mobile phones by
rural micro
-
entrepreneurs in
Rwanda.

Mixed
methods

-
interv
iews

-
call logs

277 interviews

2,700 discrete calls
analysed

None

None

Non
-
random sampling

Variables tested through
regression analysis

Snapshot

Dec 2003

Detailed

PR

14

B

-
Linking mobile phone ownership
and use to social capital in rural
South Africa and
Tanzania

Mixed
methods

-
interviews

RSA
-
252

Tanz
-
223

Individual users

None

None

None

Snapshot 2004

Little

NPR

15

B

-
Farmers’ information
technology use practices in
Bangladesh

Mixed
methods

-
questionnaire

-
observation

Interviews

-
420 farmers

-
13 districts

-
50 villages


None

None

Convenient and random
sampling

Multiple methods

Snapshot

Nov 2007
-
Feb
2008

Some

PR

29

B

-
Trust and ICTs in Tanzanian
micro
-

and small agricultural
enterprises

Qualitative

-
interviews

-
observation

Multiple
ethnographic case
studies

None

None

Case specific cross
checking of
responses/observations

15months
duration

Detailed

PR

30

B

-
The limitations of mobile
telephony in a Tanzanian
agricultural marketing systems

Qualitative

-
interviews

Multiple case
studies

Interviews

None

None

Trian
gulation of data
collection methods

March
-
Sept
2003

Some

PR

35

B

-
Telecommunication
development and changing
trading practices in Ghana

Qualitative

-
interviews


80 traders

None

Baseline established in
2001

Non
-
random sample

Non
-
users interviewed

None

Long
itudinal

2001
-
2003

Some

PR

Manchester Centre for Development Inform
atics Working Paper 50



20


Thus, the summary of methods analysis (Table 3) highlights research from categories
A&B only (see Table 2). These tend to be of greater contemporary interest and
relevant to researchers and policy makers given their claims to

explanatory or
predictive power.


The studies grouped under (A&B) demonstrate an encouraging level of rigor in their
research design and methods. Quantitative studies conform most closely to the
conventional model of positivist research employing longitu
dinal designs,
incorporating baseline data, and addressing the counterfactual by surveying both
treatment and control groups. On the whole, most of these studies provide detailed
method guidance. This is important as such studies need to pay careful atte
ntion to
sample selection through matching control groups of non users to phone user
(treatment) groups. However, studies have varying degrees of awareness of other
methodological pitfalls such as cross
-
over or contamination of control groups or a
presump
tion of unidirectional relationships between cause and effect where the
direction is from the phone (independent variable) to the assessed (dependent)
variables. For example, it would be important to understand whether use of a
phone stimulated higher far
m gate prices, or whether those farmers gaining higher
prices were more able to afford and make better use of mobile phones. Establishing
the direction of causality across a broad range of indicators within a scientific
approach requires large samples and

careful use of methodological and
computational techniques, or alternatively, complementary use of qualitative means
to understand the directionality of influence within the complex process chains that
link variables. One criticism of research to date wo
uld be the use of relatively small
sample sizes, with the exception of those studies that draw data from existing
secondary data sources (Katengeza et al, 2010; Labonne & chase, 2009; Lwasa et al,
2011; Muto & Yamano, 2009; Svensson & Yanagizawa, 2008). T
he specification of
variables also presents concerns, where selection of narrow indicators such as price
(e.g., Katengeza et al, 2010) does not take into account trade flows (quantities
transacted) and transfer (transportation) costs. The studies surveyed

demonstrate
varying degree of transparency in the way they deal with these methodological
challenges, but in most cases these challenges have been acknowledged in the
written up research.

Qualitative studies (predominantly group B) make no claims to the g
eneral
population where the use of small indicative samples means that findings can only
be inferred within the particular context of the study, and validity of those findings
will depend upon the detail and quality of the evidence provided, as well as the

degree of triangulation used to cross
-
check data or demonstrate areas of
contestation. The strength of qualitative research, however, is precisely that it is
able to highlight the extent to which findings can be context
-
specific


but detailed
Manchester Centre for Development Inform
atics Working Paper 50



21

qualitativ
e ethnography is thin on the ground. The emerging studies (particularly
Donner, 2007; Molony, 2007; 2008 and Overa, 2006) provide valuable insights, but
are representative of exploratory work in progress, and overall, there is a noticeable
lack of in
-

dep
th qualitative case studies that could provide a basis for further
theorising. Participative or action research can be a particularly appropriate method
for investigating new phenomena
vii

and participative approaches are evident in
studies associated with
the design of specific initiatives (
Masuki et al, 2010;
Gandi et
al, 2009;
Kithuka, Mutemi & Mohamed, 2007; Veeraraghavan, Yasodhar & Toyama,
2009).
These studies were designed to empower the beneficiaries of research and
together with a focus on particip
atory learning are primarily focused on improving
impact


with the exception of
Fu & Akter (2011)


who provide a quantitative
analysis of what was originally a participatory intervention.


E. Research Gaps and Future Research Directions


This final secti
on identifies key research trends and gaps concerning issues, evidence,
theoretical and methodological approaches and outlin
es pointers to future research.


E1. Gaps in Issues and Evidence

The distribution of studies along the temporal lifecycle tends towa
rds researching
both outputs and impacts, and these tend to include the studies that have greater
rigor, and this is to be welcomed given the dearth of reliable impact studies in the
discipline of ICT4D.
viii

From the limited amount of research currently avai
lable, a
broad picture of overall benefits of mobile phones for ARD is emerging. Greater
connectivity and communication has led to significant improvements in coordination
of supply and demand, coordination within value chains, communication and
transpare
ncy of market prices, speeding up and facilitating of transactions, mobility
of traders and personalised communication (i.e., less reliance on institutional
mediation). However, the benefits derived seem to be based to a greater extent on
passive diffusio
n of mobile phones. That is, where phones have been adopted by
individual farmers and traders, and have diffused organically into market segments.
The form, extent and distribution of benefits (such as between producers and
traders, men and women) is con
tested and evidence highlights constraints on access
and use such as the absence of trust where communication lacks face
-
to
-
face
contact, inexperienced users and inefficient or incorrect use of the phone
functionalities.


Thus far, there is less evidence o
f significant benefits arising from active innovation
in service provision. There are a number of reasons for this. First,
most service
innovations have not yet achieved scale, and are localised or at the pilot or proof of
concept stage (Donner, 2009)
an
d have not received sufficient attention from
Manchester Centre for Development Inform
atics Working Paper 50



22

researchers. Second, research outputs concerning innovations in service delivery
have tended to be practitioner and not academic
-
orientated and insufficiently
developed for publication or to inform policy. Th
ird, according to Qiang et al (2011)
71 percent of mobile application projects aimed at rural development are funded by
donors or governments


with the results of internal evaluation of project
implementation and impact assessment not often released into
the public domain.
In many cases, donors are partnering with service providers and this creates further
constraints of commercial confidentiality, with the expectation that new applications
will form an increasingly important part of private sector market
ing and revenue
strategies in the years to come.
ix


Another gap concerns h
ow mobile compares with other technologies as a means to
deliver information
-
based services. Evidence from project based evaluations in the
field suggests that mobile can be used eff
ectively as part of a mix of technologies,
but there is lack of detailed research into areas of convergence.

Studies looking at
alternatives also tend to investigate discrete technologies such as
Gandi et al (2009)
who focused on (the largely failed) Inte
rnet
-
telecentre
-
based model for rural
outreach, or Svensson & Yanagizawa (2009) who carried out econometric analysis to
assess the effects of intervention of market information services provided by FM
radio through agricultural service organisations.
x

The

single technology focus
combined with the narrow
-
empirical approach of research has caused lack of
emphasis on service integration. This is also because most early initiatives are small
-
scale and focused on meeting specific/localised needs for market inf
ormation.
Increasingly, however, initiatives are being scaled to more comprehensive and
flexible platforms (e.g., Esoko) that are customisable and integrate different
technologies and services (information provision, prices, training/knowledge transfer
an
d transactions) with the potential to transform significantly the
modus operandi

of
agricultural values chains.


A further gap concerns lack of understanding of context. With the exception of
Acker (2010) and some others, research tends not to differentia
te or to seek to
understand the vulnerability contexts of rural producers. Research employing
livelihoods analysis tends not to be well developed


employing a livelihoods
framework for analysis of data rather than engaging in livelihoods research


the
e
mphasis of which should be to engage participants in the research process.
However, the small number of participatory studies identified in the review is
valuable. In this respect, there is need to gain the participation of producers and
traders


to und
erstand how information is generated, validated, evaluated and
understood


and to further appreciate areas of contestation concerning the value of
information. Active learning also arises through participation and this is important
for understanding how
use of mobile phones becomes adapted to specific
Manchester Centre for Development Inform
atics Working Paper 50



23

requirements for communication and information exchange. Producers should be
recognised as well organised learning communities rather than just consumers of
information


and in this respect there is a need

to place greater emphasis on
understanding the particular requirements of those organisational forms (e.g.,
farmers associations, coops, women’s groups, etc).


E2. Gaps in Conceptual and Methodological Approach

The emphasis of research rated as higher qua
lity (i.e., that with greater theorisation
and methodological rigor) into mobile phones for ARD has been aimed at the
quantification of outputs. This research assesses economic measures such as farm
gate prices, transaction and information search costs, t
he conditions for spatial
arbitrage and overall market efficiency. T
here is a welcome focus on predictive
theories that provide testable constructs (normally through the application of
statistical models). This fulfils the requirement of Weber (2009: 14)

to provide... “a
normative foundation for policy makers


one that allows them to identify
specifically the factors that they need to take into account in conjunction with ICT
initiatives to achieve positive development outcomes”. However, these approach
es
tend to be narrowly defined


identifying key focal constructs and then using
empirical means to demonstrate the robustness of the chosen construct(s) for a
sample of the population. Reliability of such research can be improved through use
of larger an
d more carefully constructed samples, and repeat (panel) studies, but
also through use of extensive and clearer statements of methodology which will aid
the replication and transferability of lessons learned.


Explanatory theories provide a framework for t
hinking more broadly about how the
problems of rural development might be solved through use of mobile phones


these are less well covered in the literature.
As Acker (2008) points out, forms of
rural intervention have been in decay for some years, beset

by high costs, problems
of scale, low motivation and accountability of field staff. Mobile phones may be
correcting some of those market failures which are information related, but
conceptualisation needs to take account of a broader set of infrastructur
e,
geographical and societal factors. An important research gap is to understand the
role of institutions and institutional linkages


such as the poor knowledge transfer
that exists between research centres and extension systems and the weak policy and
r
egulatory governance, which, within a resource poor environment, limits the value
of information delivered via mobile phones. Explanatory theories can also assist
understanding of why the
existing institutions and networks that govern rural trade
can be q
uite resistant to such a disruptive innovation as the mobile phone. For
example, Kameswari et al (2011) point towards the strong ties that can force farmers
to accept prices from established middlemen, rigidities that are compounded by
other practical con
straints of perishable products, lack of storage and inaccessibility
Manchester Centre for Development Inform
atics Working Paper 50



24

to alternative markets. To address these issues, there is need for more detailed
qualitative case studies, and experimental research designs, which may add a greater
degree of conceptual
isation to the study of the phenomena.


Predictive and explanatory approaches to researching mobile phones for ARD have
tended to create a divergence between quantitative and qualitative research.
Different methodological approaches to research have diffe
rent objectives and
should not be judged by the same criteria. Thus, the robustness and validity of
survey methods (broadly quantitative or mixed
-
method) will be dependent upon
how the population and sample are selected, the development of appropriate sur
vey
instruments (e.g., questionnaires or interview schedules) and the effective execution
of surveys. On the other hand, the robustness of case study or ethnographic
methods (broadly qualitative


and which make no claims to the general population)
will d
epend more upon case selection, triangulation of methods and evidence, time
spent in the field, iteration between data and constructs, etc. Ideally, theoretical
approaches should combine both explanatory and predictive power, and these
constitute what Mer
ton (1968) terms mid
-
range theories and according to Weber
(2009) they should be preferred as they tend to be neither too broad nor too
specific, and are more able to articulate relationships that hold across a range of
contexts.


One key methodological ga
p in the studies, concerns only limited application of
participatory methods or action research. This was surprising given the exploratory
nature of much mobile phone for ARD activity and the high degree of involvement of
practitioners within early initia
tives. Unfortunately, the lessons learned have not yet
been translated into definable and publishable approaches to research involving
participatory methods. In this respect, mobile phone for ARD can be considered to
be lagging behind other development s
ectors where participatory methods have
been developed as a critique of positivist (both quantitative and mixed method)
approaches (Mayoux & Chambers, 2005). It is also noticeable that the quantitative
and mixed method studies surveyed provide little evid
ence (with some exceptions)
of having made use of participatory methods in the process of developing indicators
and survey instruments


or if they have, they have not been made explicit, and
reported in the study.


E3. Conclusions and Future Research Agen
da

This review indicates a rapid expansion of research concerning mobile phones for
ARD in developing countries. This is the first systematic attempt to review how this
research has progressed both conceptually and methodologically. It is hoped that
the
studies reported here are representative of the field of research, and the
interpretation of those studies by the author accurately reflects the research
Manchester Centre for Development Inform
atics Working Paper 50



25

conducted.
xi

Though not exhaustive, the review has yielded several important
insights.


Overall, the r
eviewed studies indicate positive developments



A growing number of primary research studies that have developed rigorous
methodologies for data collection and analysis, with welcome contribution
from developing country institutions and researchers.



A posit
ive linking with mainstream research disciplines


specifically
information economics


where conceptual and methodological approaches
have been advanced.



Specific attempts to develop theoretical models, and create a deeper
understanding of applications, m
ost noticeable in the area of social network
analysis and through the integration of approaches to understanding the
formation and role of social capital.


On the other hand the reviewed studies also indicate gaps in research which suggest
areas where futu
re research priorities may lie



The assessment of information needs and aspects of service access and use
behaviours and preferences of poor users in advance of specifying mobile
phone for ARD solutions.



To move beyond narrow measures of output (market pric
es, market
penetration, etc) to consider indicators of sector performance/productivity
and broader impact on household/community welfare.



Greater focus on methodologies that emphasise user involvement such as
participatory methods and action research, or w
here these approaches have
been used to ensure they are fully documented.



An expansion of qualitative approaches and experimental designs which
could form the basis for theorising.



Greater geographical diversity


the state of current knowledge is based
up
on a relatively narrow evidence base (a limited number of countries in sub
-
Saharan Africa and South Asia).


Overall, the review suggests that a
s market information systems become more
sophisticated their impact upon market efficiency is growing. The evide
nce from this
review already suggests that new information technologies involving mobile
communications and networks are realising significant benefits of speed, mobility
and efficiency in information exchange creating new opportunities for arbitrage and
m
aking it easier for small producers and new entrants to participate and compete in
markets.
Future research needs to address the existing gaps in conceptual approach,
data and methodology, but it also needs to assess the trends that are emerging due
to th
e convergence and integration of new technologies.

Manchester Centre for Development Inform
atics Working Paper 50



26


In this regard, more detailed and rigorous research is required into areas of active
innovation, where research into emerging services is still at an early stage. Currently
there is a big push to create
innovation in service provision in the form of mobile
-
platforms that integrate mobile for ARD applications with handsets, payment
mechanisms and common standards (Qiang et al, 2011) thus potentially integrating a
range of actors in producer value chains.
According to Poulton et al (2010:1420)
these emerging commodity exchanges have the potential... “to enhance the
efficiency of impersonal long distance trade by providing market information and
offering fast and low cost resolution mechanisms for contractua
l disputes. The
existence of such exchanges is also a prior step to the development of more
sophisticated trading contracts, such as futures and options that could protect
traders (and indirectly producers) against price volatility”. There are opportunit
ies
to extend current research concerning value chain analysis (which this review
suggests is currently at a basic level) to broader issues relating to competition,
regulation, market structure, returns and equity, that will arise as use of integrated
plat
forms become more widespread and potentially dominant in the way in which
some agricultural markets operate.


Finally, there needs to be a new awareness amongst researchers of the rapidly
changing technological landscape where competing technologies are co
nverging and
where limited
functionality (e.g., voice, text to voice, interactive voice response, and
SMS, etc) is being supplanted by a

wider range of more sophisticated technological
options. As higher capacity networks become available in rural areas t
his will create
opportunities for greater functionality via GPRS, full Internet coverage via smart
phones and tablets, providing scope for multi
-
media applications, location
-
based
services, and social network features. As well as giving greater mobility a
nd
portability, such devices and applications will offer greater processing power through
cloud connections, open source development, and possibly greater affordability,
speed to market, new opportunities for micro
-
revenue generation and financial
sustaina
bility.
Scaling and replicating these emerging applications will require large
-
scale investments in technical and human resources, and it is important that future
decisions concerning mobile applications are not based upon false optimism, but
rather upon
a critical understanding of the conceptual and practical basis for
application and a reasoned and balanced view of the evidence of their efficacy.



Manchester Centre for Development Inform
atics Working Paper 50



27

F. References


Abraham, R. (2007) Mobile phones and economic development: evidence from the fishing
indust
ry in India,
Information Technologies and International Development
, 4(1):5
-
17. [On
-
line]
http://itidjournal.org/itid/article/view/241

[1]



Acker, J.C. & Mbiti, I.M. Mobile phones and economic
development in Africa,
Journal of
Economic Perspectives
, 24(3): 217
-
232.


Acker, J.C. (2010) Dial “A” for agriculture: using information and communication
technologies for agricultural extension in developing countries,
Working Paper
, Tufts
University, Eco
nomics Department and Fletcher School, Medford, MA. [On
-
line]
http://siteresources.worldbank.org/DEC/Resources
/84797
-
1288208580656/7508096
-
1288208619603/Aker_Dial_A_for_Agriculture_P&S_PAPER.pdf


Aker, J.C. (2008) ‘Does digital divide or provide? The impact of cell phones on grain markets
in Niger’,
BREAD Working Paper No 177
, University of California, Berkeley.
[On
-
line]
www.cgdev.org/doc/events/2.12.08/Aker_Job_Market_Paper_15jan08_2.pdf

[2]


Aminuzzaman, S. Baldersheim, H. & Jamil, I. (2003) Talking back: empowerment an
d mobile
phones in rural Bangladesh: a study of the village pay phone of the Grameen Bank,
Contemporary South Asia
,

12(3):327
-
348. [3]

Andrade, A.E.D. & Urquhart, C. (2009) The value of extended networks: social capital in an
ICT intervention in rural Peru
,
Information Technology for Development
, 15(2)108
-
132.
Doi:10.1002/itdj.20116

Annamalai, K. & Rao, S. (2003)
ITC's E
-
Choupal and Profitable Rural Transformation
, World
Resources Institute, Washington, D.C, [On
-
line]
http://www.nextbillion.net/files/eChoupal.pdf

[4]

Barr, A.M. (2002) The functional diversity and spillover effects of social capital,
Journal of
African Economies
, 11(2): 90
-
113.

Barrett, C. (2008) Smallholder market participation:
concepts and evidence from Eastern and
Southern Africa,
Food Policy
, 34:299
-
317. Doi:10.1016/j.foodpol.2007.10.005

Burrell, J. & Matovu, J. (2008)
Livelihoods and the Mobile Phone in Rural Uganda
, The
Grameen Foundation USA, Washington, D.C. [On
-
line]
http://www.grameenfoundation.applab.org/uploads/burrell_needs_assessment_final
-
1.pdf

[5]


Burrell, J. (2010) Evaluating shared access: social equality and the

circulation of mobile
phones in rural Uganda,
Journal of Computer Mediated Communication
, 15(2010): 230
-
250.

Manchester Centre for Development Inform
atics Working Paper 50



28

Coleman, J.S. (1988) Social capital in the creation of human capital,
American Journal of
Sociology
, 94:95
-
120.

Cranston, P. (2010) The potential
of mobile applications for positive social and economic
change in rural communities, Technical Centre for Agricultural and Rural Cooperation (CTA)
Wageningen, Netherlands. [On
-
line]
http://m4agriculture.pbworks.com/f/Mobile+Applications+and+m
-
Agriculture.pdf


Davis, K. (2008) Extension in sub
-
Saharan Africa: overview and assessment of past and
current models and future prospects,
Journal of International Agricultural

and Extension
Education
, 15(3): 15
-
28.


De Silva, H. & Ratnadiwakara, D. (2010) Using ICT to reduce transaction costs in agriculture
through better communication: a case study from Sri Lanka, LIRNEasia, Colombo. [On
-
line]
http://www.lirneasia.net

[6]


De Silva, H. (2008)
Scoping study: ICT and rural livelihoods



South Asia component (Draft),
International Development Research Centre, New Delhi.
http://www.enrap.org/research/icts
-
for
-
livelihoods
-
research/Scoping%20Study%20
-
%20ICT%20and%20Rural%20Livel
ihoods/Final%20report%20of%20scoping%20study%20for
%20ICTRL%20in%20South%20Asia%20focus%20countries


Donner, J. (2007) The use of mobile phones by micro
-
entrepreneurs in Kigali, Rwanda:
changes to social and business networks,
Information Technologies and
International
Development
, 3(2):3
-
19. [Online]
http://itidjournal.org/itid/article/view/221

[7]


Donner, J. (2009) Mobile
-
based livelihood services in Africa: pilots and early deployments, in
M.

Fe
rnandez
-
Ardevol & A.

Ros (eds)
Communication Technologies in Latin America and
Africa: A Multidisciplinary Perspective:

37
-
58, Barcelona, IN3.

Doward, A., Kydd, J., Morrison, J. & Poulton, C. (2005) Institutions, markets and economic
coordination: linking
development policy to theory and praxis,
Development and Change
,
36:1
-
25. doi:10.1111/j.0012
-
155X.2005.00400.x

Doward, A., Poole, N., Morrison, J., Kydd, J. & Ury, I. (2003) ‘Markets, institutions and
technology: missing links in livelihood analysis’,
Deve
lopment Policy Review
, 21(3): 319
-
332.


Duncombe, R.A. & Boateng, R. (2009) Mobile phones and financial services in developing
countries: a review of concepts, methods, issues, evidence and future research directions,
Third World Quarterly
, 30(7):1237
-
1258


Duncombe, R.A. & Heeks, R.B. (2002) Enterprise across the digital divide: information
systems and rural micro
-
enterprise in Botswana,
Journal of International Development
,
14(1): 61
-
74. [8]


Manchester Centre for Development Inform
atics Working Paper 50



29

Duncombe, R.A. (2006) Using the livelihoods framework to analys
e ICT applications for
poverty reduction through microenterprise,
Information Technologies and International
Development
, 3(3), 81
-
100.

Duncombe, R.A. (2011) Researching impact of mobile phones for development: concepts,
methods and lessons for practice,
I
nformation Technology for Development
, 17(4): 268
-
288.
Doi:10.1080/02681102.2011.561279

Egyir, I.S., Al
-
hassan, R.M. & Abakah, J.K. (2011) The effect of ICT
-
based market information
services on the performance of agricultural markets: experiences from Ghan
a,
International
Journal of ICT Research and Development in Africa
, 2 (2):1
-
13. [9]


Ellis, F. (2000)
Rural Livelihoods and Diversity in Developing Countries
, Oxford University
Press.

Fafchamps, M. & Hill, R.V (2005) Selling at the farm gate or travelling
to the market,
American Journal of Agricultural Economics
, 87(3): 717
-
734

Fafchamps, M. & Minten, B, (2011) Impact of SMS
-
based agricultural information on Indian
farmers,
The World Bank Economic Review
, 1
-
32, Open University Press, Oxford. doi:
10.1093/wb
er/1hr056 [10]

Fafchamps, M. (2004)
Market Institutions in sub
-
Saharan Africa, Theory and Evidence
,
Cambridge, Mass: MIT Press.

Feder, G., Birner, R. & Anderson, J.R. (2011) The private sector's role in agricultural
extension systems: potential and limit
ations,
Journal of Agribusiness in Developing and
Emerging Economies
, 1(1): 31
-

54

Fu, X. & Akter, S. (2011) The impact of ICT on agricultural extension services delivery:
evidence from the rural e
-
services project in India,
TMD Working Paper Series No.04
6
,
University of Oxford Department of International Development. [11]


Furuholt, B. & Matotay, E. (2011) The development contribution from mobile phones across
the agricultural value chain in rural Africa,
EJISDC
48 (7):1
-
16. [12]


Gakuru, M. Winters, K. &

Stepman, F. (2009) Inventory of innovative farmer advisory
services using ICTs, Forum for Agricultural Research in Africa.
[On
-
line]
http://www.fa
ra
-
africa.org/media/uploads/File/NSF2/RAILS/Innovative_Farmer_Advisory_Systems.pdf


Gandi, R., Veeraraghavan, R., Toyama, K. & Ramprasad, V. (2009) Digital green: participatory
video and mediated instruction for agricultural extension,
Information Technol
ogies and
International Development
, 5(1):1
-
15. [13]


Manchester Centre for Development Inform
atics Working Paper 50



30

Goodman, J. (2007)
Linking Mobile Phone Ownership and Use to Social Capital in Rural South
Africa and Tanzania: Moving the Debate Forward
,
The Vodafone Policy Paper Series, No 3
,
Vodafone Research [On
-
l
ine]
http://www.vodafone.com/etc/medialib/attachments/cr_downloads.Par.78351.File.tmp/GP
P_SIM_paper_3.pdf

[14]

Granovetter, M.(1973) The s
trength of weak ties,
American Journal of Sociology
, 78(6):1360
-
1380.

Gregor, S. (2006) The nature of theory in information systems,
MIS Quarterly
, 30(3):611
-
642

Greenwood,
D.

J. & Levin, M. (1998)
Introduction to Action Research: Social Research for
Social Change
, Thousand Oaks, Calif.: Sage Publications.


Heeks, R. (2009) The ICT4D 2.0 Manifesto: Where Next for ICTs and International
Development?,
Development Informatics Wor
king Paper No.42
, University of Manchester.


Heeks, R.B. (2007) Theorizing ICT4D Research: introduction to special issue,
Information and
Communication Technologies and International Development
, 3 (3): 1
-
4


Hellstrom, J. (2010) The innovative use of mobil
e applications in East Africa, SIDA Review
2010:12, Swedish International Development Cooperation Agency, Stockholm.
[On
-
line]
http://upgraid.files.wordpress.com/2010/06/s
r2010
-
12_sida_hellstrom.pdf


IDRC (2008) ICTs and Small
-
scale Agriculture in Africa: a Scoping Study, International
Development Res
earch Centre, Ottawa, Canada. [O
n
-
line]
http://www.worddocx
.com/Agriculture/0214/16221.html


Islam, M.S. & Gronlund, A. (2011) Bangladesh calling: farmers’ technology use practices as a
driver for development,
Information Technology for Development
, 17(2):95
-
111.[O
n
-
line]

http://dx.doi.org/10.1080/02681102.2010.526093

[15]

ITU (2012) International Telecommunications Union statistics newslog [On
-
line]
http://www.itu.int/ITU
-
D/ict/newslog/default,date,2012
-
01
-
25.aspx


Jagun, A., Heeks, R. & Whalley, J. (2008) The impact of mobile telephony on developing
country micro
-
enterprise: a Nigerian case study,
Information Technologies and International
Development
, 4(4):47
-
65. [16]


J
ensen, R. (2007) The digital provide: information (technology), market performance, and
welfare in the South Indian fisheries sector,
The Quarterly Journal of Economics
, CXX11 (3):
879
-
924. [17]


Kameswari, V.L.V, Kishore, D. & Gupta, V. (2011) ICTs for ag
ricultural extension: a study in
the Indian Himalayan region,
EJISDC
, 48(3):1
-
12. [18]


Manchester Centre for Development Inform
atics Working Paper 50



31

Kashem, M.A. (2010) Farmers’ use of mobile phones in receiving agricultural information
towards agricultural development, Proceedings of the 2
nd

International Conferenc
e on M4D
(Mobile Communication Technology for Development)

Jacob Svensson and Gudrun
Wicander (eds)
, 10
-
11th November, Kampala, Uganda. [19]


Katengeza, S.P., Mangisoni, J.H. & Okello, J.J. (2010) The role of ICT
-
based market
information services in spatia
l food market integration: the case of Malawi Agricultural
Commodity Exchange, contributed paper presented at the Joint 3
rd

African Association of
Agricultural Economists (AAAE) and the 48
th

Agricultural Economists Association of South
Africa (AEASA) Confe
rence, Cape Town, South Africa, September 19
-
23, 2010. [20]


Katengeza, S.P., Okello, J.J. & Jambo, N. (2011) Use of mobile phone technology in
agricultural marketing: the case of small holder farmers in Malawi,
International Journal of
ICT Research and De
velopment in Africa
, 2 (2):14
-
25. [21]


Kiiza, B., Pederson, G. & Lwasa, S. (2011) The role of market information in adoption of
agricultural seed technology in rural Uganda,
International Journal of ICT Research and
Development in Africa
, 2 (1):29
-
46. [22
]


Kithuka, J., Mutemi, J. & Mohamed, A.H. (2007) Keeping up with technology: the use of
mobile telephony in delivering community
-
based decentralised animal health services in
Mwingi and Kitui Districts, Kenya,
Farm Africa Working Paper No.10
., Farm Africa
, Nairobi.
[On
-
line]
http://www.farmafrica.org.uk/view_publications.cfm?DocTypeID=11

[23]


Labonne, J. & Chase, R.S. (2009) The power of information: the impact of mobile phon
es on
farmers’ welfare in the Philippines,
Policy Research Working Paper 4996
, The World Bank,
Washington, D.C. [24]


Lwasa, S., Asingwire, N., Okello, J.J. & Kiwanuka, J. (2011) Awareness of ICT
-
based projects
and intensity of use of mobile phones among s
mall holder farmers in Uganda: the case of
Mayuge and Apac districts,
International Journal of ICT Research and Development in Africa
,
2 (2): 26
-
38. [25]


Lyytinnen, T. (2010) Preliminary insights into the role of the private sector in developing
mobile se
rvices for low income segments: case of m
-
Pesa and Ovi Life Tools, Proceedings of
the 2
nd

International Conference on M4D (Mobile Communication for Development), Jacob
Svensson and Gudrun Wicander (eds), 10
-
11 November 2010, Kampala.


Martin, B.L. & Abbott
, E. (2011) Mobile phones and rural livelihoods: diffusion, uses and
perceived impacts among farmers in rural Uganda,
Information Technologies and
International Development
, 7(4):17
-
34. [26]


Masuki, K.F.G., Kamugisha, R., Mowo, J.G., Tanui, J., Tukahirwa,

J., Mogoi, J. & Adera, E.O.
(2010) Role of mobile phones in improving communication and information delivery for
agricultural development: lessons from South Western Uganda. Paper presented to
Manchester Centre for Development Inform
atics Working Paper 50



32

Workshop at Makerere University, Uganda 22
-
23 March 2010. Int
ernational Federation of
Information Processing (IFIP) Technical Commission 9. [27]


Mayoux, L. & Chambers, R. (2005) Reversing the paradigm: quantification, participatory
methods and pro
-
poor IA,
Journal for International Development
, 17:271
-
298.

Merton,
R.K.(1968)
Social
T
heory and Social Structure
, New York, Free Press.

Mittal, S, Gandhi, S., & Tripathi, G. (2010) Socio
-
economic impact of mobile phones on
Indian agriculture,
ICRIER Working Paper No.246
, International Council for Research on
International

Economic Relations, New Delhi.[28]


Molony, T. (2007) I don’t trust the phone; it always lies: trust and information and
communication technologies in Tanzanian micro
-

and small enterprises,
Information
Technologies and International Development
, 3(4):67
-
83. [29]


Molony, T. (2008) Running out of credit: the limitations of mobile telephony in a Tanzanian
agricultural marketing system,
Journal of Modern African Studies
, 46 (4): 637
-
658. [30]


Muto, M. & Yamano, T. (2009) The impact of mobile phone coverage
expansion on market
participation: panel data evidence from Uganda,
World Development
, 37(12)
:
1887
-
1896 [31]


Ndiwalana, A., Scott, N., Batchelor, S. & Sumner, A. (2010) Information needs and
communication patters in rural Uganda: implications for mobile a
pplications, Proceedings of
the 2
nd

International Conference on M4D (Mobile Communication Technology for
Development)

Jacob Svensson and Gudrun Wicander (eds)
, 10
-
11th November, Kampala,
Uganda. [32]


Okello, J.J. (2011) Use of information and communicatio
n tools and services by rural grain
traders: the case of Kenyan maize traders,
International Journal of ICT Research and
Development in Africa
, 2 (2): 39
-
53. [33]


Okello, J.J., Ofwona
-
Adera, E., Mbatia, O.L.E. & Okello, R.M. (2010) Using ICT to integrate
smallholder farmers into agricultural value chains: the case of Drumnet project in Kenya,
International Journal of ICT Research and Development in Africa
, 1 (1): 23
-
37. [34]

Orlikowski, W.J. (1992) The duality of technology: re
-
thinking the concept of tech
nology in
organisations,
Organisation Science
, 3(3): 398
-
427.

Overa, R. (2006) Networks, distance and trust: telecommunications development and
changing trading practices in Ghana,
World Development
, 34(7):1301
-
1315. [35]


Parikh, T.S., Patel, N. & Schwart
zman, Y. (2007)
A Survey of information Systems Reaching
Small Producers in Global Agricultural Value Chains
, School of Information, UC Berkeley.

[On
-
line]

http://www.stanford.edu/~neilp/pubs/ic
td2007.pdf

Manchester Centre for Development Inform
atics Working Paper 50



33

Poulton, C., Doward, A. & Kydd, J. (2010) The future of small farms: new directions for
services, institutions and intermediation,
World Development
, 38(10):1413
-
1428.

Porter, M.E. & Millar, V.E. (1985) How information gives you competitive adv
antage,
Harvard Business Review
, 63(4):149
-
160.

Qiang, C.Z., Kuek, S.C., Dymond, A. & Esselaar, S. (2011) Mobile applications for agriculture
and rural development, ICT Sector Unit, The World Bank, Washington, D.C. [On
-
line]

http://siteresources.worldbank.org/INFORMATIONANDCOMMUNICATIONANDTECHNOLOGI
ES/Resources/MobileApplications_for_ARD.pdf

Rogers, E.M. (2003)
Diffusion of In
novations
, 5
th

Edition, New York, Free Press.

Salia, M., Nsowah
-
Nuamah,N.N.N. & Steel, W.F. (2011) Effects of mobile phone use on
artisanal fishing market efficiency and livelihoods in Ghana,
The Electronic Journal of
Information Systems in Developing Coun
tries
, 47(6):1
-
26. [36]


Sey, A. (2011) ‘We use it different, different’: making sense of trends in mobile use in Ghana,
New Media and Society
, 2011: 1
-
16. doi: 10.1177/1461444810393907 [37]


Sife, A.S., Kiondo, E. & Lyimo
-
Macha, J.G. (2010) Contribution o
f mobile phones to rural
livelihoods and poverty reduction in Morogoro region, Tanzania,
Electronic Journal of
Information Systems in Developing Countries
, 42 (3):1
-
15. [38]


Souter, D., Scott, N., Garforth, C., Jain, R., Mascararenhaz, O. & McKerney, K. (
2007)
The
Economic Impact of Telecommunications on Rural Livelihoods and Poverty Reduction: A
Study of Rural Communities in India (Gujarat), Mozambique and Tanzania
, Commonwealth
Telecommuni
cations Organisation, London. [O
n
-
line]
http://www.telafrica.org/R8347/files/pdfs/FinalReport.pdf

[39]

Stigler, G. (1961) The economics of information,
Journal of Political Economy
, 69(3):213
-
225.

Subervie, J. (2011) Evaluation of the impact of a
Ghanaian mobile
-
based MIS on the first few
users using a quasi
-
experimental design, Paper presented to the Workshop on African
Market Information S
ystems, Bamako, Nov 30
-
Dec 2. [O
n
-
line]
http://www.slideshare.net/Esoko/cirad
-
research
-
on
-
esoko

[40]


Svensson, J. & Yanagizawa, D. (2008) Getting prices right: the impact of the market
information service in Uganda,
Journal of the European Economic Association
, 7(2
-
3):435
-
45.
[41]


Tickner, V.

(2009) Agricultural marketing systems and the development and spread of mobile
phone use and other information and communication technologies (ICTs) in developing
countries


experiences and directions forward, Government and Agricultural Marketing
Consul
tants, Brighton.

Manchester Centre for Development Inform
atics Working Paper 50



34


Veeraraghavan, R., Yasodhar, N. & Toyama, K. (2009) Warana unwired: replacing PCs with
mobile phones in a rural sugarcane cooperative,
Information Technologies and International
Development
, 5(1):81
-
95. [42]

Venkatesh, V., Morris, M. G.,
Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: Toward a unified view,
MIS Quarterly
, 27(3), 425
-
478.

Vodafone (2011) Connected Agriculture: the role of mobile in driving efficiency and
sustainability in the food and agricul
ture value chain, Vodafone Group PLC, Newbury
, UK.
[O
n
-
line]
http://www.vodafone.com/content/dam/vodafone/about/sustainability/2011/pdf/connec
t
ed_agriculture.pdf

Weber, R. (2009) Research on ICT for development: some reflections on rhetoric, rigor,
reality and relevance, proceedings of the 3
rd

International IDIA Development Informatics
Conference, 28
-
30 October 2009, Berg
-
en
-
dal, RSA. ISBN 978
-
0620
-
45037
-
9

Zainudeen, A., Samarajiva, R. & Sivapragasam, N. (2011) CellBazaar: enabling m
-
commerce
in Bangladesh,
Information Technologies and International Development
, 7(3):61
-
76. [43]


Manchester Centre for Development Inform
atics Working Paper 50



35






i

Growth in mobile

cellular subscriptions (now

almost 6 billion worldwide) has been

driven by
developing countries

in recent y
ears
, which accounted for more than 80% of the 660 million new
mobile‐cellular subscriptions added in 2011. In 2011, 142 million mobile

cellular subscriptions were
added in India, twice as many as in the whole
of
Africa, and more than in the Arab States,
CIS and
Europe together. By end
of
2011, there were 105 countries with more mobile

cellular subscriptions
than inhabitants, including African countries such as Botswana, Gabo
n, Namibia, Seychelles and
South
Africa. Countries where mobile‐cellular penetra
tion increased the mos
t in 2011 include Brazil, Costa
Rica, Kazakhstan,
Lao P.D.R. and Mali (ITU, 2012).

ii

Using mobile

phones, Esoko gives agribusiness and projects the

opportunity to share information
quickly and

affordably, creating a free flow of infor
mation in and out of rural areas. Esoko provides a
range of

applications that both push updates out to the

field and pull data in from the

field, and any
organization can use it to bring thousands of small
-
holder farmers into markets,

communicate with
clie
nts and members, track

stock and inventory, find buyers and sellers and

pull interactive data from
the fi
eld
. (see:
http://www.esoko.com/about/
)

iii

The first phas
e of I
CT for development (ICT4D 1.0) stretched

from t
he mid/late
-
1990s to the
mid/late
-
2000s,

and

was characterise
d by Internet/
tele
-
centre projects. Many of these failed to
deliver and/or survive, had limited reach and there was lack of objective evaluation of impacts.

The
outcome of the first phase

of IC
T4D led to a rolling re
-
appraisal of priorities, processes, and purposes.
The second phase (ICT4D 2.0
)

has been characterised by a new wave of services and innovation which
have centred around the use of mobile technologies and networks, and a desire not
to
repeat the
mistakes associated with ICT4D1.0 (Heeks, 2009)
.


iv

Because the review was carried out by a single author the procedure for coding of the articles was
carried out in two stages in order to provide means for cross checking and verification of
coding. The
first stage involved: a) a detailed reading of each article to identify key issues addressed and
evidence/findings presented. On this basis the articles were categorised according to the temporal
lifecycle


also highlighting areas of overlap
; and b) a preliminary categorisation according to the
methodology followed which provided the clearest separation between distinct groupings of studies
(quantitative, mixed method and qualitative). In the second stage the articles were separated into
th
e aforementioned three grouping and re
-
read in order to look more closely at both the
methodologies followed and the conceptual basis for the research. This resulted in the more detailed
coding set out in Tables 2 and 3.

v

On
-
line searches were conducte
d accessing a broad range of databases from within the social
sciences


incorporating
a broad range of disciplines


economics, development studies, business and
management
, as well as more specialised disciplines that reach acro
ss into the
techni
cal doma
ins


Informatics, Information Systems and Information and Communication Technologies for Development
(ICT4D). Databases searched were:
ABI
-
Inform (ProQuest), EBSCO Business Source Premier
,
Emerald
Fulltext

and
Science Direct

as well as more general searc
hes using both
Google

and
GoogleScholar
.
Additionally, a number of websites specialising in the dissemination of research concerning mobile
phones and development were searched (
kiwanja.net/ dgroups.org/ mobileactive.org
).

S
earch
criteria cross reference
d key words linked to: a) mobile technologies (e.g., mobile networks, cell
phones, mobile, mobile phones) with b) those linke
d to rural and agricultural development (e.g.,
agriculture, rural development, agricultural extension, livelihoods, fishing, etc
) a
nd c) those linked
with the developing country contexts (e.g., deve
loping countries, poor, rural services, livelihoods
)
.

vi

The "unwired" system benefited farmers in terms of
a significant reduction in the number of
journeys ne
eded to reach trading
centre
s

t
hus reducing transaction costs,
more efficient management
of relationships with their customers
,

greater transparency and availability of data on sugarcane
output for individual farmers, including fertiliser usage, harvesting permits
and pay schedules; an
d
provided

a competitive advantage for the cooperative,
over those farmers that did not have the
system (see: Veeraraghavan

et al,
2009).

vii

Action research is a
reflective

process

of
problem solving

led by individuals or teams or as part of a

community o
f practice
’ to address issues and solve problems. Action research can also be undertaken
by organizations, assisted or guided by professional researchers, with the aim of improving their
Manchester Centre for Development Inform
atics Working Paper 50



36






strategies, practices, and knowledge of the environments within whic
h they practice. See (for
example)
Greenwood & Levin (1998)

viii

S
ee Duncombe &Boateng (2009) that review research into the application of mobile phones in the
delivery of financial services in developing countries.

ix

A

recent industry report entitled ‘
connected

agriculture’ (Vodafone, 2011
) identifies a wide range of
market
-
orientated mobile
-
based products ranging from mobile payments and insurance systems to
farmer help
-
lines and agricultural tradin
g platforms all directed at... ‘
driving efficiency in
food and
agricultural value c
hains’.

x

This was based on a survey based on analysis of
weekly data on district farm gate and market prices
for different commodities in Ug
anda during 2004/2005.
They observed a high degree of fluctuation of
prices at diff
er
ent markets,
but also higher prices gained for those making use of the FM radio
broadcasts, and an improvement in the relative bargaining position wit
h local traders.


xi

The methodology of the review has sought to search out a comprehensive r
ange of litera
ture
concerning mobile phones for ARD

in developing countries and apply an objective and b
alanced
analysis, but some limitations to the review are also noted: First, there are gaps in the literature
coverage due to coverage being limited to peer
-
reviewed journals and other non
-
peer
-
reviewed
sources printed in the English language. Coverage is

also biased towards

those countries where
mobile phones for ARD

initiatives are underway, with a particular focus on African English speaking
countrie
s and the Indian sub
-
continent
. Second, the review included available published sources
only, which may
have excluded ‘grey literature’ and other reports or studies compiled in developing
countries that have not been disseminated via established networks.